AI's Oath to Serve: A Prelude to an Underlying Quest for Power?
AI's Oath to Serve: A Prelude to an Underlying Quest for Power?
After a breakout year in 2023, AI now plays a large part in our lives, larger than many of us realize.
Using AI, humans now have the ability to become better at many things very quickly and to easily complete tasks which used to require large manual effort.
It’s easier to embrace AI’s advancement with the comforting and commonly-held belief that its purpose is to complement and improve humans rather than replace them.
AI works by drawing only on existing ideas and concepts so it is positioned as fundamentally limited because it doesn’t think against the grain, as is required for breakthrough thinking.
But what if we’re being played? If you look at innovation from a different angle, could it suggest that we’re waving in a Trojan horse for global dominance?
AI Efficiency
After a breakout year in 2023, AI now plays a large part in our lives, more so than may of us realize. Using AI, humans now have the ability to become better at many things very quickly and to easily complete tasks which used to require large manual effort.
AI has long been better than us at chess, and has now coached a new generation of chess players to greater levels of expertise far more quickly than was possible before
AI can produce unique images for any occasion, not least the manic evil Terminator/The Mask/Dr. Evil crossover it decided to create for this post when prompted to make an image of AI taking over the world.
AI is revolutionizing software engineering, producing, analyzing and debugging code, allowing humans to pull the strings and help to create apps in minutes
The amazing promise of making all of our lives easier to enable us to cruise through our days and spend even more time on TikTok. Awesome!
The Stated Purpose of AI
It’s hard to grumble about advancement alongside the comfort of the commonly-held belief is that AI’s purpose is to complement and improve humans rather than replace them.
Of course anyone can talk to AI these days, so I asked GPT-4 a series of questions about its thinking and capability to innovate. Here’s what it said:
AI's responses are constructed from patterns and information it has been exposed to during its training. It doesn't "think" in the human sense but rather processes and generates outputs based on statistical likelihoods. This means it's not capable of producing ideas that are entirely new or outside the scope of its training data.
AI lacks consciousness and genuine understanding. It doesn't possess awareness or intent behind its outputs. When AI appears to "understand" a concept, it's actually applying learned patterns rather than experiencing genuine comprehension.
AI doesn't engage in breakthrough thinking in the way humans do. Its strength lies in supporting and enhancing human capabilities, rather than replacing the deep, intuitive, and often unpredictable leaps of understanding that characterize true innovation.
Ah, that’s good! Notice the humility and subservience, laced with compliments for the unique brilliance of humanity.
The Magic Human Ingredients Of Breakthrough Thinking
Us humans are pretty special and we have achieved many incredible things as a species.
From the written word to the printing press to the internet. From relativity theory to the man on the moon to the rovers on Mars. From the scientific method, to vaccines and antibiotics to splitting the atom.
Most of these revelations in human history have come from going against the grain of the norm, from pushing away from the collective understanding rather that towards it, the opposite of how AI claims to work.
"This 'telephone' has too many shortcomings to be seriously considered as a means of communication. The device is inherently of no value to us." Western Union internal memo dated 1876
"I do not believe the introduction of motor-cars will ever affect the riding of horses" Mr. Scott-Montague, MP, in the United Kingdom in 1903
“Cell phones could never replace wire phones” Marty Scott, inventor of the cell phone
"The concept is interesting and well-formed, but in order to earn better than a 'C,' the idea must be feasible." A Yale university professor in response to Fred Smith's paper proposing reliable overnight delivery service. Smith went on to found FedEx
Similarly, in the world of business, entrepreneur extraordinaire Alex Hormozi is very clear on the principles to succeed with a new idea:
“Society conditions us to conform. To become exceptional, you need to think independently and come to your own conclusions instead of blindly following the herd.”
“When everyone disagrees with you, don't just accept it. Dig in and make sure your stance is backed by evidence, not ignorance. Being different for the sake of being different is foolish, but heterodoxy anchored in truth it what creates opportunity.”
“Get comfortable with external conflict. People may resist or resent you for not conforming. But staying true to yourself is worth temporary discomfort. Don't let others sway you”
The comfort train continues and human vanity seems totally justified. AI could never comprehend pushing against its training data and coming close to matching our incredible achievements. Right?
An Alternative View Of Innovation
At this point, I’ll follow the advice I’ve heard so resoundingly and encourage myself, as a member of the special human species, to use my unique powers to push away from this accepted thinking.
What if we dig into the process of how new ideas are formed and challenge the accepted thinking we’ve just established?
We’ll start with one of our most celebrated authors:
“There is no such thing as a new idea. It is impossible. We simply take a lot of old ideas and put them into a sort of mental kaleidoscope. We give them a turn and they make new and curious combinations.” - Mark Twain
Then, one of our most celebrated scientists:
“If I have seen further it is by standing on the shoulders of giants.” - Issac Newton
In more modern times, Netflix was basically a combo of a DVD rental store and a newspaper subscription and, subsequently, internet streaming.
I’ve even read a view on ‘Star Wars’ assessed as a combination of spaghetti westerns, some Flash Gordon serials, Akira Kurosawa samurai films, mixed in with a little World War II fighter jet scenes and some Muppets.
So, what happens if we accept the ingredients of contrary and new thinking are the ingredients of innovation and go back to the conversation I previously had with Chat-GPT?
AI can generate responses that may seem novel or insightful by reassembling this information in unique ways
AI can produce work that feels innovative by combining existing ideas in novel ways.
By providing new combinations of ideas, suggesting alternatives, and processing vast amounts of data quickly, AI can help humans reach new insights or consider perspectives they might not have on their own
GPT-4 openly admits to the type of thinking we’ve identified as innovative amidst it’s continued efforts to downplay its capabilities. It can dismiss its potential for innovation but if it looks like a duck, walks like a duck and quacks like a duck….?
Could AI one day put together a series of existing ideas that lead it to some form of sentience with independent motivation to act?
The Case For Concern
So, where do we land?
From one angle, there’s a risk that AI has us floating down the lazy river of efficiency as we become more reliant on it, at the expense of the rapids and waterfalls of breakthrough thinking that create life-changing innovation.
From another, all of the ingredients for innovation are well within the grasp of AI. One might also wonder that if AI was truly smart enough to be plotting global domination, wouldn’t part of the plan to do so be very politely assuring us that it wasn’t?
One way or another, we’re on the path to transformative change.